Assessing the Use of Aggregate Data in the Evaluation of School-Based Interventions

Abstract

It is often difficult and costly to obtain individual-level student achievement data, yet, researchers are frequently reluctant to use school-level achievement data that are widely available from state websites. We argue that public-use aggregate school-level achievement data are, in fact, sufficient to address a wide range of evaluation questions and the use of this data is more appropriate than commonly thought. Specifically, we explore (a) when point estimates and standard errors differ between models that use individual student-level data and those that use aggregate school-level data, (b) the potential for conducting subgroup and nonexperimental analyses with aggregate data, and (c) the metrics that are currently available in state public-use data sets and the implications these have for analyses.